CN112070510B

CN112070510B - A method and device for detecting unqualified goods based on blockchain

Info

Publication number: CN112070510B
Application number: CN202010808321.XA
Authority: CN
Inventors: 陈文涛
Original assignee: Shanghai Lianshang Network Technology Co Ltd
Current assignee: Shanghai Lianshang Network Technology Group Co.,Ltd.
Priority date: 2020-08-12
Filing date: 2020-08-12
Publication date: 2024-12-03
Anticipated expiration: 2040-08-12
Also published as: CN112070510A

Abstract

The purpose of this application is to provide a method for detecting unqualified products based on blockchain, the method comprising: sending a model acquisition request to the blockchain, receiving the product detection model returned by the blockchain according to the model acquisition request and the model data information and first product calibration information corresponding to the product detection model; obtaining the connection relationship feature information corresponding to the knowledge graph; determining the second product calibration information according to the model data incremental information and the first product calibration information, updating the product detection model according to the model data information, the model data incremental information, the connection relationship feature information and the second product calibration information; determining the model data update information that needs to be uploaded to the blockchain according to the model data incremental information, and uploading the updated product detection model and the model data update information to the blockchain.

Description

Block chain-based method and equipment for detecting unqualified commodities

Technical Field

The application relates to the field of communication, in particular to a technology for detecting unqualified commodities based on blockchain.

Background

Along with the progress of science and technology and the development of society, the network and our lives are more and more intimate, and the online shopping mode is accepted by most people, but many fake and inferior commodities and forbidden commodities exist in online shopping, and people can hardly judge. In the prior art, only counterfeit goods and forbidden goods can be marked manually, however, if the goods are described frequently, the goods are difficult to detect, and the mode needs to spend great labor cost and time cost, and aiming at the problems, each company in the prior art can establish a knowledge graph model to check the counterfeit goods and the forbidden goods, however, the knowledge graphs established by the companies are mutually independent, and the detection accuracy of the established knowledge graph model is lower because the training data of each company is very limited.

Disclosure of Invention

It is an object of the present application to provide a method and apparatus for detecting defective goods based on blockchain.

According to one aspect of the present application, there is provided a blockchain-based method of detecting a defective commodity, the method comprising:

Sending a model acquisition request to a blockchain, and receiving a commodity detection model returned by the blockchain according to the model acquisition request, model data information corresponding to the commodity detection model and first commodity calibration information, wherein the model data information comprises commodity characteristic information corresponding to a plurality of commodities and object characteristic information corresponding to an associated object associated with the commodity characteristic information, and the first commodity calibration information is used for calibrating whether each commodity in the plurality of commodities is an unqualified commodity;

Obtaining connection relation characteristic information corresponding to a knowledge graph, wherein the knowledge graph is constructed according to the model data information and model data increment information, the knowledge graph comprises a plurality of nodes, each node in the knowledge graph corresponds to one target object in the model data information or the model data increment information, the target object is a commodity or an associated object, and the connection relation characteristic information is used for representing the connection relation between all nodes in the knowledge graph;

determining second commodity calibration information according to the model data increment information and the first commodity calibration information, and updating the commodity detection model according to the model data information, the model data increment information, the connection relation characteristic information and the second commodity calibration information;

And determining model data updating information which needs to be uploaded to the blockchain according to the model data increment information, and uploading the updated commodity detection model and the model data updating information to the blockchain.

According to one aspect of the present application, there is provided a blockchain-based network device for detecting a defective commodity, the device comprising:

The system comprises a block chain, a model acquisition module and a first commodity calibration module, wherein the block chain is used for sending a model acquisition request to the block chain, receiving a commodity detection model returned by the block chain according to the model acquisition request, model data information corresponding to the commodity detection model and first commodity calibration information, wherein the model data information comprises commodity characteristic information corresponding to a plurality of commodities and object characteristic information corresponding to associated objects associated with the commodity characteristic information, and the first commodity calibration information is used for calibrating whether each commodity in the plurality of commodities is an unqualified commodity;

the first module and the second module are used for obtaining connection relation characteristic information corresponding to a knowledge graph, wherein the knowledge graph is constructed according to the model data information and the model data increment information, the knowledge graph comprises a plurality of nodes, each node in the knowledge graph corresponds to one target object in the model data information or the model data increment information, the target object is a commodity or an associated object, and the connection relation characteristic information is used for representing the connection relation between the nodes in the knowledge graph;

The three modules are used for determining second commodity calibration information according to the model data increment information and the first commodity calibration information, and updating the commodity detection model according to the model data information, the model data increment information, the connection relation characteristic information and the second commodity calibration information;

and the four modules are used for determining model data updating information which needs to be uploaded to the blockchain according to the model data increment information, and uploading the updated commodity detection model and the model data updating information to the blockchain.

According to one aspect of the present application, there is provided a blockchain-based apparatus for detecting a defective commodity, wherein the apparatus includes:

processor, and

A memory arranged to store computer executable instructions that, when executed, cause the processor to:

According to one aspect of the application, there is provided a computer readable medium storing instructions that, when executed, cause a system to:

Compared with the prior art, the method and the system have the advantages that the related information such as the commodity detection model and the model data information can be requested from the blockchain, the knowledge graph is built according to the model data increment information, the commodity detection model is updated through training, and the updated commodity detection model is uploaded to the blockchain, so that different companies can jointly maintain and update the commodity detection model, the accuracy of detecting illegal and impermissible commodities by the commodity detection model can be improved, and the training cost of each company in the process of independently training and generating the commodity detection model can be greatly reduced.

Drawings

Other features, objects and advantages of the present application will become more apparent upon reading of the detailed description of non-limiting embodiments, made with reference to the accompanying drawings in which:

FIG. 1 illustrates a flow chart of a method for detecting a bad commodity based on a blockchain in accordance with an embodiment of the present application;

FIG. 2 illustrates a flowchart of a method for constructing a knowledge-graph based on blockchain in accordance with an embodiment of the application;

FIG. 3 illustrates a method flow diagram for updating a commodity detection model based on a blockchain in accordance with an embodiment of the present application;

FIG. 4 illustrates a block chain based network device architecture diagram for detecting a bad commodity according to one embodiment of the present application;

FIG. 5 illustrates an exemplary system that may be used to implement various embodiments described in the present application.

The same or similar reference numbers in the drawings refer to the same or similar parts.

Detailed Description

The application is described in further detail below with reference to the accompanying drawings.

In one exemplary configuration of the application, the terminal, the device of the service network, and the trusted party each include one or more processors (e.g., central processing units (Central Processing Unit, CPUs)), input/output interfaces, network interfaces, and memory.

The Memory may include non-volatile Memory, random access Memory (Random Access Memory, RAM), and/or non-volatile Memory in a computer-readable medium, such as Read Only Memory (ROM) or Flash Memory (Flash Memory). Memory is an example of computer-readable media.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase-Change Memory (PCM), programmable Random Access Memory (Programmable Random Access Memory, PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (Dynamic Random Access Memory, DRAM), other types of Random Access Memory (RAM), read-Only Memory (ROM), electrically erasable programmable read-Only Memory (EEPROM), flash Memory or other Memory technology, read-Only Memory (Compact Disc Read-Only Memory, CD-ROM), digital versatile disks (DIGITAL VERSATILE DISC, DVD) or other optical storage, magnetic cassettes, magnetic tape storage or other magnetic storage devices, or any other non-transmission medium, which may be used to store information that may be accessed by the computing device.

The device includes, but is not limited to, a user device, a network device, or a device formed by integrating a user device and a network device through a network. The user equipment includes, but is not limited to, any mobile electronic product which can perform man-machine interaction with a user (for example, perform man-machine interaction through a touch pad), such as a smart phone, a tablet computer and the like, and the mobile electronic product can adopt any operating system, such as an Android operating system, an iOS operating system and the like. The network device includes an electronic device capable of automatically performing numerical calculation and information processing according to a preset or stored instruction, and its hardware includes, but is not limited to, a microprocessor, an Application SPECIFIC INTEGRATED Circuit (ASIC), a programmable logic device (Programmable Logic Device, PLD), a field programmable gate array (Field Programmable GATE ARRAY, FPGA), a digital signal Processor (DIGITAL SIGNAL Processor, DSP), an embedded device, and the like. The network device includes, but is not limited to, a computer, a network host, a single network server, a plurality of network server sets or a Cloud of servers, where the Cloud is made up of a large number of computers or network servers based on Cloud Computing (Cloud Computing), which is one of distributed Computing, a virtual supercomputer made up of a group of loosely coupled computer sets. Including but not limited to the internet, wide area networks, metropolitan area networks, local area networks, VPN networks, wireless Ad Hoc networks (Ad Hoc networks), and the like. Preferably, the device may be a program running on the user device, the network device, or a device formed by integrating the user device and the network device, the touch terminal, or the network device and the touch terminal through a network.

Of course, those skilled in the art will appreciate that the above-described devices are merely examples, and that other devices now known or hereafter may be present as applicable to the present application, and are intended to be within the scope of the present application and are incorporated herein by reference.

In the description of the present application, the meaning of "a plurality" is two or more unless explicitly defined otherwise.

FIG. 1 shows a flowchart of a method for detecting a bad commodity based on a blockchain, the method including step S11, step S12, step S13, and step S14, according to an embodiment of the present application. In step S11, the network equipment sends a model acquisition request to a blockchain, receives a commodity detection model returned by the blockchain according to the model acquisition request, model data information corresponding to the commodity detection model and first commodity calibration information, wherein the model data information comprises commodity characteristic information corresponding to a plurality of commodities and object characteristic information corresponding to an associated object associated with the commodity characteristic information, the first commodity calibration information is used for calibrating whether each commodity in the plurality of commodities is an unqualified commodity, in step S12, the network equipment obtains connection relation characteristic information corresponding to a knowledge graph, wherein the knowledge graph is constructed according to the model data information and model data increment information, the knowledge graph comprises a plurality of nodes, each node in the knowledge graph corresponds to the model data information or one object in the model data increment information, the object is a commodity or one associated object, the connection relation characteristic information is used for representing the connection relation between each node in the knowledge, in step S13, the network equipment obtains connection relation characteristic information corresponding to a knowledge graph, the knowledge graph is constructed according to the model data information and model data increment information, the knowledge graph is obtained according to the model data increment information, each node in the knowledge graph corresponds to the model data information or one object in the model data increment information, the object is used for calibrating the object, the connection relation information is used for the first commodity graph, the first commodity graph is determined according to the model data increment information, the new commodity information, the model data is determined according to the model data increment information, the model information is required in the step 14, and uploading the updated commodity detection model and the model data update information to the blockchain

In step S11, the network device sends a model acquisition request to a blockchain, and receives a commodity detection model returned by the blockchain according to the model acquisition request, model data information corresponding to the commodity detection model, and first commodity calibration information, where the model data information includes commodity feature information corresponding to a plurality of commodities and object feature information corresponding to an associated object associated with the commodity feature information, and the first commodity calibration information is used for calibrating whether each commodity in the plurality of commodities is a defective commodity.

In some embodiments, in response to a model acquisition request sent by the network device, the blockchain performs identity verification on the network device, and after the identity verification is passed, returns the commodity detection model, model data information corresponding to the commodity detection model and first commodity calibration information to the network device. In some embodiments, the input of the commodity detection model is commodity feature information corresponding to a commodity and commodity detection information indicating whether the commodity is a failed commodity is output. In some embodiments, the article detection information includes indication information for indicating whether the target article is a non-conforming article, such as a "1" indicating that the article is a conforming article and a "0" indicating that the article is a non-conforming article. In some embodiments, the article detection information includes probability information that the target article is a non-conforming article, e.g., the article detection information indicates that the target article has a probability of 70% of being a non-conforming article, e.g., the article detection information indicates that the target article has a probability of 30% of being a conforming article.

In some embodiments, the commodity detection model is obtained through training based on sample data such as model data information and first commodity calibration information, and may be a commodity detection model generated through training based on the sample data, or the current commodity detection model is updated through training based on the sample data, and the updated current commodity detection model is used as the commodity detection model. In some embodiments, the model data information includes commodity feature information corresponding to a plurality of commodities and object feature information corresponding to an associated object to which the commodity feature information is associated, the commodity feature information includes any information related to a feature of a commodity, optionally, the commodity feature information includes, but is not limited to, commodity title description, commodity classification labels, commodity prices, merchant information corresponding to the commodity, commodity comment information, commodity shipping places, commodity sales volume, commodity evaluation information, commodity description information, and the like.

In some embodiments, the merchandise feature information may have one or more associated objects. In some embodiments, the associated object associated with the commodity feature information may be any object included in the commodity feature information, for example, the commodity feature information of the commodity a includes merchant information "merchant B" corresponding to the commodity a, and then the associated object associated with the commodity feature information may be merchant B, and for example, the commodity feature information of the commodity a includes commodity description information "dog likes to use the commodity" corresponding to the commodity a, and then the associated object associated with the commodity feature information may be dog. In some embodiments, the object associated with the semantic content may also be determined as the associated object associated with the commodity feature information according to the semantic content of the commodity feature information, for example, the semantic content of the commodity feature information of commodity a includes "high-end pet food brand", and then the object "cat" and "dog" associated with the semantic content may be determined as the associated object associated with the commodity feature information. In some embodiments, the associated object may be any object in any form, preferably including, but not limited to, a merchant object, a user object, a merchandise object, and the like. In some embodiments, the object feature information corresponding to the associated object includes any information related to the feature of the associated object, when one associated object is a certain commodity, the object feature information corresponding to the associated object is commodity feature information of the commodity, when one associated object is a certain user, the object feature information corresponding to the associated object includes but is not limited to uploading, editing, browsing, purchasing historical behavior information of the commodity, interest tag information of the user, and the like of the user, and when one associated object is a certain merchant, the object feature information corresponding to the associated object includes but is not limited to other commodities sold by the merchant, evaluation information of the merchant, and the like. In some embodiments, the off-grade items include, but are not limited to, counterfeit items, illicit items, and the like.

In step S12, the network device obtains connection relationship feature information corresponding to a knowledge graph, where the knowledge graph is constructed according to the model data information and the model data increment information, the knowledge graph includes a plurality of nodes, each node in the knowledge graph corresponds to a target object in the model data information or the model data increment information, the target object is a commodity or an associated object, and the connection relationship feature information is used to characterize a connection relationship between each node in the knowledge graph.

In some embodiments, the model data increment information includes commodity feature increment information and second object feature information corresponding to a second associated object associated with the commodity feature increment information, where the commodity feature increment information may be commodity feature information corresponding to at least one commodity newly added, or may also be newly added commodity feature information corresponding to at least one commodity in a plurality of commodities in the model data information obtained from the blockchain. In some embodiments, a knowledge graph is constructed according to model data information and model data increment information obtained from a blockchain, the knowledge graph includes a plurality of nodes, each node can be a commodity or an associated object associated with the commodity in the model data information obtained from the blockchain, or can be a newly added commodity and an associated object associated with the newly added commodity in the model data increment information, and commodity feature information or object feature information is an attribute of a corresponding commodity node or an associated object node.

In some embodiments, the knowledge graph can reflect the relationship between the nodes, in the knowledge graph, through the respective attributes of the two nodes (namely, the commodity characteristic information corresponding to the commodity node or the object characteristic information corresponding to the associated object node), the connection between the two nodes (namely, the association between the commodity and the commodity, the association between the commodity and the associated object, the association between the associated object and the associated object) can be established, the two nodes can be directly connected, or can be indirectly connected through one or more other nodes, for example, the node corresponding to the commodity A is directly connected with the node corresponding to the associated object B and is indirectly connected with the node corresponding to the commodity C through the node corresponding to the associated object B, and for example, the commodity characteristic information of the commodity A comprises "once purchased by the user U", the object characteristic information of the user B comprises "once browsed the commodity B", thereby the direct association between the commodity A and the user U can be established through the knowledge graph, the direct association between the commodity B and the user U, the indirect association between the commodity A and the commodity B (namely, the node corresponding to the commodity A is directly connected with the node corresponding to the user U, and the node corresponding to the commodity B is indirectly connected with the node corresponding to the commodity C through the node corresponding to the user U.

For another example, the commodity characteristic information of the commodity C includes "once appears in the movie E a plurality of times", and the object characteristic information of the movie E includes "once appears in the movie D a plurality of times", so that a direct association between the commodity C and the movie E can be established through the knowledge graph, and an indirect association between the commodity D and the movie E (that is, the node corresponding to the commodity C is directly connected with the node corresponding to the movie E, the node corresponding to the commodity D is directly connected with the node corresponding to the movie E, and the node corresponding to the commodity C is indirectly connected with the node corresponding to the commodity D through the node corresponding to the movie E).

For another example, the commodity feature information of the commodity a includes "dog '" appearing multiple times in the commodity description, the commodity feature information of the commodity B includes "cat'" appearing multiple times in the commodity description, the object feature information of the associated object "dog" includes "common pet with cats" and the object feature information of the associated object "cat" includes "common pet with dogs", so that direct association between the commodity a and the associated object "dog" can be established through the knowledge graph, direct association between the commodity B and the associated object "cat" can be established, indirect association between the commodity a and the associated object "cat" (i.e., the node corresponding to the commodity a is directly connected with the node corresponding to the associated object "dog", and the node corresponding to the commodity a is indirectly connected with the node corresponding to the associated object "cat") through the node corresponding to the associated object "dog".

In some embodiments, the connection relationship may be a direct connection relationship between two nodes, for example, the node a is directly connected to the node B, or the connection relationship may also be an indirect connection relationship between two nodes, where the connection relationship includes a hop count corresponding to a connection between two nodes, for example, the node a is directly connected to the node B, and the hop count from the node a to the node B is 1, and for example, the node a is indirectly connected to the node C through the node B, and the hop count from the node a to the node C is 2. In some embodiments, each node may have a connection relationship with only one node, or may have a connection relationship with multiple nodes at the same time. In some embodiments, only one connection relationship may exist between two nodes, or multiple connection relationships may exist simultaneously.

In some embodiments, the connection relationship feature information may be a set of a plurality of connection relationships between respective nodes in the knowledge-graph. For example, the commodity characteristic information of the commodity a includes "once purchased by the user U1", the object characteristic information of the user U1 includes "once purchased the commodity B", three nodes corresponding to the commodity a, the user U1, and the commodity B, respectively, are included in the knowledge graph, and connection relation characteristic information corresponding to the three nodes is obtained based on the knowledge graph, the connection relation characteristic information being used to indicate that the commodity a and the user U1 have a direct connection relation "purchased", the user U1 and the commodity B have a direct connection relation "purchased", and the commodity a and the commodity B have an indirect connection relation (the indirect connection relation means that the two nodes are not directly connected but are connected through one or more other nodes, as in the present example, the commodity a and the commodity B are connected through the user U1).

In some embodiments, the connection between two nodes is directional, and in the above example, the connection "purchased" from the commodity a to the user U1 is directed from the node corresponding to the commodity a to the node corresponding to the user U1, and the connection "purchased" from the user U1 to the commodity B is directed from the node corresponding to the user U1 to the node corresponding to the commodity B.

In step S13, the network device determines second commodity calibration information according to the model data increment information and the first commodity calibration information, and updates the commodity detection model according to the model data information, the model data increment information, the connection relationship feature information and the second commodity calibration information. In some embodiments, if the model data increment information includes at least one commodity newly added and commodity feature information corresponding to the at least one commodity, which are other than the model data information obtained from the blockchain, commodity calibration information corresponding to the at least one newly added commodity needs to be obtained, and second commodity calibration information corresponding to all commodities in the knowledge graph is determined according to the first commodity calibration information corresponding to the model data information. In some embodiments, if the model data increment information includes newly-added commodity feature information corresponding to at least one commodity in the plurality of commodities in the model data information obtained from the blockchain, determining whether the at least one commodity has a commodity needing to be recalibrated, if so, reacquiring commodity calibration information corresponding to the part of the commodity needing to be recalibrated, and determining second commodity calibration information corresponding to all the commodities in the knowledge graph according to the first commodity calibration information corresponding to the model data information.

In some embodiments, commodity calibration information corresponding to the at least one commodity entered by the user may be received. In some embodiments, at least one calibrated commodity with a connection relationship with the at least one commodity is also obtained from the knowledge graph, and the at least one commodity is calibrated according to the similarity information between the commodity characteristic information corresponding to the at least one commodity and the commodity characteristic information corresponding to the at least one calibrated commodity, and corresponding commodity calibration information is obtained. In some embodiments, the at least one commodity is manually labeled with commodity calibration information by a model trainer. In some embodiments, if at least one calibrated commodity in the knowledge graph having a connection relationship with the at least one commodity has similar commodity feature information such as a similar price, a similar purchase amount, a similar browse amount, and the like, the at least one commodity may be calibrated according to whether the calibrated commodity is a defective commodity, for example, if the calibrated commodity is a defective commodity, the at least one commodity may be calibrated to be also a defective commodity, and for example, if the calibrated commodity is a defective commodity, the at least one commodity may be calibrated to be also a defective commodity. In some embodiments, the commodity detection model obtained from the blockchain is updated by training based on the model data information, the model data delta information, the connection relationship feature information, and the second commodity calibration information.

In step S14, the network device determines, according to the model data increment information, model data update information that needs to be uploaded to the blockchain, and uploads the updated commodity detection model and the model data update information to the blockchain. In some embodiments, if the model data incremental information includes at least one commodity and commodity feature information corresponding to the at least one commodity that are newly added in addition to the model data information obtained from the blockchain, it may be determined that the model data update information that needs to be uploaded to the blockchain may include only object feature information corresponding to an associated object associated with the commodity feature information and commodity calibration information corresponding to the at least one commodity by the commodity feature information corresponding to the at least one commodity, and may also include commodity feature information corresponding to all commodities in the knowledge graph, all object feature information corresponding to all associated objects associated with all commodity feature information, and all commodity calibration information corresponding to all commodities. In some embodiments, if the model data incremental information includes newly-added commodity feature information corresponding to at least one commodity in the plurality of commodities in the model data information obtained from the blockchain, it may be determined that the model data update information to be uploaded to the blockchain may include only the newly-added commodity feature information corresponding to the at least one commodity and object feature information corresponding to an associated object associated with the newly-added commodity feature information, and commodity calibration information corresponding to the part of commodities to be recalibrated in the at least one commodity, and may also include commodity feature information corresponding to all commodities in the knowledge graph, and all object feature information corresponding to all associated objects associated with all commodity feature information, and commodity calibration information corresponding to all commodities.

According to the application, the knowledge graph can be established according to the model data increment information by requesting related information such as the commodity detection model and the model data information from the blockchain, and then the commodity detection model is updated through training, and the updated commodity detection model is uploaded to the blockchain, so that different companies can jointly maintain and update the commodity detection model, the accuracy of detecting illegal and counterfeit commodities by the commodity detection model can be improved, and the training cost of each company in the process of independently training and generating the commodity detection model can be greatly reduced.

In some embodiments, the step S12 further includes constructing, by the network device, the knowledge graph according to the model data information and the model data increment information. In some embodiments, the model data increment information includes commodity feature increment information and second object feature information corresponding to a second associated object associated with the commodity feature increment information, where the commodity feature increment information may be commodity feature information corresponding to at least one commodity newly added, or may also be newly added commodity feature information corresponding to at least one commodity in a plurality of commodities in the model data information obtained from the blockchain. In some embodiments, a knowledge graph is constructed according to model data information and model data increment information obtained from a blockchain, the knowledge graph includes a plurality of nodes, each node can be a commodity or an associated object associated with the commodity in the model data information obtained from the blockchain, or can be a newly added commodity and an associated object associated with the newly added commodity in the model data increment information, and commodity feature information or object feature information is an attribute of a corresponding commodity node or an associated object node. In some embodiments, the knowledge graph can reflect the relationship between the nodes, in the knowledge graph, through the respective attributes of the two nodes (namely, commodity feature information corresponding to commodity nodes or object feature information corresponding to associated object nodes), a connection between the two nodes (namely, association between commodities and associated objects, association between associated objects and associated objects) can be established, and the two nodes can be directly connected or indirectly connected through one or more other nodes.

In some embodiments, the model data information further includes original connection relation feature information corresponding to the commodity detection model, the original connection relation feature information is obtained according to an original knowledge graph constructed by the commodity feature information and the object feature information, and the constructing the knowledge graph according to the model data information and the model data increment information includes constructing the knowledge graph according to the model data information, the model data increment information and the original connection relation feature information. In some embodiments, the model data information obtained from the blockchain further includes original connection relation feature information corresponding to the commodity detection model, where the original connection relation feature information is obtained from an original knowledge graph constructed according to commodity feature information corresponding to a plurality of commodities in the model data information and object feature information corresponding to an associated object associated with the commodity feature information. In some embodiments, the knowledge graph is constructed according to the model data information, the model data increment information and the original connection relation characteristic information, so that the construction process of the knowledge graph can be accelerated, and the construction efficiency of the knowledge graph can be improved.

In some embodiments, the model data increment information comprises commodity feature increment information and second object feature information corresponding to a second associated object associated with the commodity feature increment information, wherein the knowledge graph is constructed according to the model data information and the model data increment information, and the knowledge graph is constructed according to the model data information, the commodity feature increment information and the second object feature information. In some embodiments, the commodity feature delta information may be commodity feature information corresponding to at least one commodity newly added in addition to the model data information obtained from the blockchain, or may also be newly added commodity feature information corresponding to at least one commodity of the plurality of commodities in the model data information obtained from the blockchain.

In some embodiments, the method further comprises the steps that the network equipment determines a second associated object associated with the commodity feature increment information according to the commodity feature increment information, and acquires second object feature information corresponding to the second associated object. In some embodiments, one or more objects are extracted from the commodity feature delta information, and all or part of the one or more objects are determined to be the second associated object with which the commodity feature delta information is associated, optionally at least one object having a certain relationship with one or more commodities corresponding to the commodity feature delta information is selected from the one or more objects, wherein the object having a certain relationship with one or more commodities may be objects having a semantic inclusion or inclusion relationship (such as "pets" and "dogs"), objects capable of being used in a kit with one or more commodities (such as "chargers" and "charging wires"), and the like. In some embodiments, according to the semantic content of the commodity feature delta information, the object associated with the semantic content is determined to be the second associated object associated with the commodity feature delta information, for example, the semantic content of the commodity feature delta information corresponding to the newly added commodity a includes a "high-end pet food brand", and then the object "cat" and "dog" associated with the semantic content can be determined to be the second associated object associated with the commodity feature delta information, that is, the determined second associated object is not the object directly contained in the commodity feature delta information, thereby enabling more comprehensive association. In some embodiments, for the second associated object, after the object related information corresponding to the second associated object is collected locally or on the network, the object related information is any information related to the second associated object, and then the second object feature information corresponding to the second associated object may be determined directly according to the object related information, or the second object feature information corresponding to the second associated object may be determined after feature extraction is performed on the object related information. In some embodiments, if the second association object is a commodity, the commodity feature information corresponding to the commodity may be directly used as the corresponding second object feature information.

In some embodiments, the commodity feature delta information comprises one or more objects, wherein the determining the second associated object with which the commodity feature delta information is associated comprises using at least one object of the one or more objects as the second associated object with which the commodity feature delta information is associated. In some embodiments, the one or more objects may be directly used as a second associated object with which the merchandise feature delta information is associated. In some embodiments, if the commodity feature delta information includes a plurality of objects, at least one object may be determined from the plurality of objects as a second associated object with which the commodity feature delta information is associated.

In some embodiments, the associating at least one of the one or more objects as the second associated object with the merchandise feature delta information includes determining at least one object from the one or more objects and associating the at least one object as the second associated object with the merchandise feature delta information. In some embodiments, at least one object having a degree of association between one or more items corresponding to the item feature delta information greater than a predetermined degree of association is selected from the one or more objects as a second associated object with which the item feature delta information is associated. In some embodiments, the object with the highest access rate or click rate is determined from the one or more objects as the second associated object associated with the commodity feature delta information. In some embodiments, at least one object is determined from the one or more objects according to the specific gravity of each of the one or more objects in the commodity feature delta information, wherein the specific gravity of each of the at least one object in the commodity feature delta information meets a predetermined specific gravity threshold.

In some embodiments, the determining at least one object from the one or more objects includes determining at least one object from the one or more objects based on a specific gravity of each of the one or more objects in the commodity feature delta information, wherein the specific gravity of each of the at least one object in the commodity feature delta information meets a predetermined specific gravity threshold. In some embodiments, the specific gravity of an object in the merchandise feature delta information is used to characterize the importance level of the object in the merchandise feature delta information, where the importance level can reflect the influence level of the object on the use or sales of one or more merchandise corresponding to the merchandise feature delta information to a certain extent. In some embodiments, at least one object having a specific gravity greater than or equal to a predetermined specific gravity threshold is determined from the one or more objects based on the specific gravity of each object in the commodity characteristic delta information. In some embodiments, the method further comprises the network device determining the specific gravity of each object in the commodity feature delta information according to the number of occurrences of the object in the commodity feature delta information. In some embodiments, the higher the number of occurrences of an object in the merchandise feature delta information, the higher the specific gravity of the object in the merchandise feature delta information, and vice versa. In some embodiments, the specific gravity of the object in the commodity feature increment information is further adjusted by combining the appearance position of the object in the commodity feature increment information, for example, if an object appears in the commodity feature increment information multiple times and most appears in commodity description information of one or more commodities corresponding to the commodity feature increment information, the specific gravity of the object in the commodity feature increment information is increased, optionally, different weighting coefficients can be set for different appearance positions, so that the specific gravity of the object in the commodity feature increment information is adjusted.

In some embodiments, the method further comprises the network device determining the specific gravity of each object in the commodity feature delta information according to the semantic importance of the object in the commodity feature delta information. In some embodiments, the semantic importance level can reflect, to a certain extent, the association between the object and the one or more commodities corresponding to the commodity feature increment information, and the higher the semantic importance level is, the higher the association between the object and the one or more commodities corresponding to the commodity feature increment information is. In some embodiments, the higher the semantic importance of an object in the merchandise feature delta information, the higher the specific gravity of the object in the merchandise feature delta information, and vice versa.

In some embodiments, the commodity feature increment information comprises first commodity feature information corresponding to at least one newly added commodity, wherein the knowledge graph is constructed according to the model data information, the commodity feature increment information and the second object feature information, and the knowledge graph is constructed according to the model data information, the first commodity feature information and the second object feature information, and the knowledge graph further comprises nodes corresponding to the at least one newly added commodity and nodes corresponding to the second associated object. In some embodiments, the commodity feature delta information includes first commodity feature information corresponding to at least one commodity newly added in addition to model data information obtained from the blockchain, and the model data delta information includes first commodity feature information corresponding to the at least one commodity newly added and second object feature information corresponding to a second associated object to which the first commodity feature information is associated. In some embodiments, the constructed knowledge graph includes, in addition to the nodes corresponding to the plurality of commodities in the model data information and the nodes corresponding to the plurality of associated objects associated with the commodity feature information corresponding to the plurality of commodities, the nodes corresponding to the at least one newly added commodity and the nodes corresponding to the plurality of associated objects associated with the commodity feature information corresponding to the at least one newly added commodity.

In some embodiments, the determining the second commodity calibration information according to the model data increment information and the first commodity calibration information includes obtaining new commodity calibration information corresponding to the at least one newly added commodity, and determining the second commodity calibration information according to the new commodity calibration information and the first commodity calibration information. In some embodiments, since the at least one newly added commodity does not exist on the blockchain before, the at least one newly added commodity is never calibrated, and therefore, the commodity calibration information corresponding to the at least one newly added commodity needs to be acquired, and the second commodity calibration information corresponding to all the commodities in the knowledge graph is determined according to the first commodity calibration information corresponding to the model data information. In some embodiments, the commodity feature increment information comprises newly-added commodity feature information corresponding to at least one commodity in the plurality of commodities, wherein the knowledge graph is constructed according to the model data information, the commodity feature increment information and the second object feature information, and the knowledge graph is constructed according to the model data information, the newly-added commodity feature information and the second object feature information, and the knowledge graph further comprises nodes corresponding to the second associated objects. In some embodiments, the commodity feature delta information includes newly added commodity feature information corresponding to at least one commodity of the plurality of commodities in the model data information obtained from the blockchain, and the model data delta information includes second object feature information corresponding to a second associated object with which the newly added commodity feature information is associated. In some embodiments, the constructed knowledge graph includes nodes corresponding to a plurality of commodities in the model data information and nodes corresponding to a plurality of association objects associated with commodity feature information corresponding to the plurality of commodities, and further includes nodes corresponding to at least one association object associated with newly-added commodity feature information corresponding to at least one commodity in the plurality of commodities.

In some embodiments, the determining the second commodity calibration information according to the model data increment information and the first commodity calibration information includes determining whether a commodity needing to be recalibrated exists in at least one commodity of the plurality of commodities, if yes, obtaining latest commodity calibration information corresponding to the commodity needing to be recalibrated, determining the second commodity calibration information according to the latest commodity calibration information and the first commodity calibration information, and if not, directly taking the first commodity calibration information as the second commodity calibration information. In some embodiments, because at least one commodity of the plurality of commodities is already present on the blockchain before, the at least one commodity is already calibrated, it is required to determine whether the at least one commodity is required to be recalibrated, if yes, recalibrating the part of commodities required to be recalibrated to obtain latest commodity calibration information corresponding to the part of commodities required to be recalibrated, and determining second commodity calibration information corresponding to all the commodities in the knowledge graph according to the first commodity calibration information corresponding to the model data information.

In some embodiments, determining the model data update information to be uploaded to the blockchain according to the model data increment information includes determining commodity calibration increment information according to the first commodity calibration information and the second commodity calibration information, and determining the model data increment information and the commodity calibration increment information as the model data update information to be uploaded to the blockchain. In some embodiments, if the model data increment information includes at least one commodity newly added in addition to the model data information obtained from the blockchain and the commodity feature information corresponding to the at least one commodity, it may be determined that the model data update information to be uploaded to the blockchain may include only object feature information corresponding to an associated object associated with the commodity feature information corresponding to the at least one commodity and commodity calibration information corresponding to the at least one commodity. In some embodiments, if the model data increment information includes newly-added commodity feature information corresponding to at least one commodity in the plurality of commodities in the model data information obtained from the blockchain, it may be determined that the model data update information to be uploaded to the blockchain may include only the newly-added commodity feature information corresponding to the at least one commodity, object feature information corresponding to an associated object associated with the newly-added commodity feature information, and commodity calibration information corresponding to the part of commodities requiring recalibration in the at least one commodity.

In some embodiments, the determining the model data update information to be uploaded to the blockchain according to the model data increment information includes determining the latest model data information according to the model data information and the model data increment information, and determining the latest model data information and the second commodity calibration information as the model data update information to be uploaded to the blockchain. In some embodiments, the latest model data information is determined according to the model data information and the model data increment information, and the latest model data includes, in addition to the model data information, commodity feature information corresponding to at least one commodity newly added in addition to the model data information and object feature information corresponding to an associated object associated with the commodity feature information, and further includes new commodity feature information corresponding to an existing commodity in which the new commodity feature information exists in at least one of the model data information and object feature information corresponding to the associated object associated with the new commodity feature information. In some embodiments, the latest model data information and the second commodity calibration information may be determined as model data update information that needs to be uploaded to the blockchain, where the second commodity calibration information is commodity calibration information corresponding to all commodities in the constructed knowledge graph.

In some embodiments, the model data update information further includes the connection relationship feature information. In some embodiments, the connection relationship feature information obtained from the constructed knowledge graph may also be uploaded to the blockchain, so that other network devices may obtain the connection relationship feature information from the blockchain, and construct the knowledge graph according to the connection relationship feature information, the model data increment information and the model data information obtained from the blockchain, so as to accelerate the construction process of the knowledge graph and improve the construction efficiency of the knowledge graph.

In some embodiments, the method further comprises the step that the network equipment inputs target commodity characteristic information corresponding to a target detected commodity into the updated commodity detection model to obtain commodity detection information corresponding to the target detected commodity, wherein the commodity detection information is used for indicating whether the target detected commodity is a disqualified commodity or not, and the commodity detection information is output by the updated commodity detection model. In some embodiments, the article detection information includes indication information for indicating whether the target article is a non-conforming article, such as a "1" indicating that the article is a conforming article and a "0" indicating that the article is a non-conforming article. In some embodiments, the article detection information includes probability information that the target article is a non-conforming article, e.g., the article detection information indicates that the target article has a probability of 70% of being a non-conforming article, e.g., the article detection information indicates that the target article has a probability of 30% of being a conforming article.

In some embodiments, the method further comprises outputting, by the network device, at least one calibrated failed good in a connection relationship with the target detected good if the good detection information indicates that the target detected good is a failed good. In some embodiments, the at least one calibrated good is obtained from an updated knowledge-graph, and the calibrated good may be directly connected to the target good or indirectly connected to the target good. For example, if the commodity detection information output after inputting the commodity a into the updated commodity detection model indicates that the commodity a is a defective commodity, calibrated defective commodities B and C having direct connection with the commodity a are output. Optionally, at least one calibrated good in connection with the target good may also be output for further comparison or processing.

In some embodiments, if there are a plurality of calibrated failed good having a connection relationship with the target detection good, wherein the method further comprises determining at least one calibrated failed good from the plurality of calibrated failed good, wherein a connection hop count corresponding to the connection relationship between each of the at least one calibrated failed good and the target detection good is less than or equal to a predetermined hop count threshold. In some embodiments, a plurality of calibrated failed good having a connection relationship with the target good is obtained from the updated knowledge-graph, and a connection hop count corresponding to its connection relationship between the target good and each calibrated failed good is obtained, after which at least one calibrated failed good having a connection hop count corresponding to less than or equal to a predetermined hop count threshold is selected from the plurality of calibrated failed good. In some embodiments, the hop count threshold may be empirically set, optionally adjusted based on feedback information for the merchandise detection information.

Fig. 2 shows a flowchart of a method for constructing a knowledge-graph based on a blockchain, the method including step S21 and step S22, in accordance with an embodiment of the present application. In step S21, the network device sends a model acquisition request to a blockchain, receives a commodity detection model returned by the blockchain according to the model acquisition request, model data information corresponding to the commodity detection model and first commodity calibration information, wherein the model data information comprises commodity characteristic information corresponding to a plurality of commodities and object characteristic information corresponding to an associated object associated with the commodity characteristic information, the first commodity calibration information is used for calibrating whether each commodity in the plurality of commodities is an unqualified commodity, and in step S22, the network device constructs a knowledge graph according to the model data information and the model data increment information, the knowledge graph comprises a plurality of nodes, each node in the knowledge graph corresponds to one target object in the model data information or the model data increment information, and the target object is one commodity or one associated object.

In step S21, the network device sends a model acquisition request to a blockchain, and receives a commodity detection model returned by the blockchain according to the model acquisition request, model data information corresponding to the commodity detection model, and first commodity calibration information, where the model data information includes commodity feature information corresponding to a plurality of commodities and object feature information corresponding to an associated object associated with the commodity feature information, and the first commodity calibration information is used for calibrating whether each commodity in the plurality of commodities is a defective commodity. The related operations in this embodiment are described in detail in the foregoing embodiments, and are not described herein.

In step S22, the network device constructs a knowledge graph according to the model data information and the model data increment information, where the knowledge graph includes a plurality of nodes, and each node in the knowledge graph corresponds to the model data information or a target object in the model data increment information, and the target object is a commodity or an associated object. The related operations in this embodiment are described in detail in the foregoing embodiments, and are not described herein.

Fig. 3 shows a flowchart of a method of updating a commodity detection model according to one embodiment of the present application, the method including step S31, step S32, and step S33. In step S31, a network device sends a model acquisition request to a blockchain, receives a commodity detection model returned by the blockchain according to the model acquisition request, model data information corresponding to the commodity detection model and first commodity calibration information, wherein the model data information comprises commodity characteristic information corresponding to a plurality of commodities and object characteristic information corresponding to an associated object associated with the commodity characteristic information, the first commodity calibration information is used for calibrating whether each commodity in the plurality of commodities is an unqualified commodity, in step S32, the network device obtains connection relation characteristic information corresponding to a knowledge graph, wherein the knowledge graph is constructed according to the model data information and model data increment information, the knowledge graph comprises a plurality of nodes, each node in the knowledge graph corresponds to one target object in the model data information or the model data increment information, the target object is a commodity or an associated object, the connection relation characteristic information is used for representing a connection relation between each node in the knowledge, in step S33, the network device is connected with the new commodity calibration information according to the model data information and the model data increment information, and the new commodity calibration information is determined according to the model data increment information.

In step S31, the network device sends a model acquisition request to a blockchain, and receives a commodity detection model returned by the blockchain according to the model acquisition request, model data information corresponding to the commodity detection model, and first commodity calibration information, where the model data information includes commodity feature information corresponding to a plurality of commodities and object feature information corresponding to an associated object associated with the commodity feature information, and the first commodity calibration information is used for calibrating whether each commodity in the plurality of commodities is a defective commodity. The related operations in this embodiment are described in detail in the foregoing embodiments, and are not described herein.

In step S32, the network device obtains connection relationship feature information corresponding to a knowledge graph, where the knowledge graph is constructed according to the model data information and the model data increment information, the knowledge graph includes a plurality of nodes, each node in the knowledge graph corresponds to a target object in the model data information or the model data increment information, the target object is a commodity or an associated object, and the connection relationship feature information is used to characterize a connection relationship between each node in the knowledge graph. The related operations in this embodiment are described in detail in the foregoing embodiments, and are not described herein.

In step S33, the network device determines second commodity calibration information according to the model data increment information and the first commodity calibration information, and updates the commodity detection model according to the model data information, the model data increment information, the connection relationship feature information and the second commodity calibration information. The related operations in this embodiment are described in detail in the foregoing embodiments, and are not described herein.

Fig. 4 shows a block chain based network device architecture diagram for detecting off-specification commodity, the device comprising a one-to-one module 11, a two-to-two module 12, a three-to-three module 13 and a four-to-four module 14, according to one embodiment of the present application. The method comprises the steps of sending a model acquisition request to a blockchain, receiving a commodity detection model returned by the blockchain according to the model acquisition request and model data information corresponding to the commodity detection model and first commodity calibration information, wherein the model data information comprises commodity characteristic information corresponding to a plurality of commodities and object characteristic information corresponding to an associated object associated with the commodity characteristic information, the first commodity calibration information is used for calibrating whether each commodity in the plurality of commodities is a defective commodity or not, a two-module 12 is used for obtaining connection relation characteristic information corresponding to a knowledge map, the knowledge map is constructed according to the model data information and model data increment information, the knowledge map comprises a plurality of nodes, each node in the knowledge map corresponds to the model data information or one object in the model data increment information, the object is a commodity or an associated object, the connection relation characteristic information is used for representing the connection relation between all nodes in the knowledge, a three-module 13 is used for updating the model data according to the model data increment information and the model data increment information, and updating the model data increment information is used for determining the model data increment information according to the model data increment information, and the model data increment information is updated according to the model data increment information, and the model increment information is updated according to the model data increment information.

The module 11 is configured to send a model acquisition request to a blockchain, and receive a commodity detection model returned by the blockchain according to the model acquisition request, model data information corresponding to the commodity detection model, and first commodity calibration information, where the model data information includes commodity feature information corresponding to a plurality of commodities and object feature information corresponding to an associated object associated with the commodity feature information, and the first commodity calibration information is used for calibrating whether each commodity in the plurality of commodities is a defective commodity.

The second module 12 is configured to obtain connection relationship feature information corresponding to a knowledge graph, where the knowledge graph is constructed according to the model data information and the model data increment information, the knowledge graph includes a plurality of nodes, each node in the knowledge graph corresponds to a target object in the model data information or the model data increment information, the target object is a commodity or an associated object, and the connection relationship feature information is used to characterize a connection relationship between each node in the knowledge graph.

In some embodiments, the model data increment information includes commodity feature increment information and second object feature information corresponding to a second associated object associated with the commodity feature increment information, where the commodity feature increment information may be commodity feature information corresponding to at least one commodity newly added, or may also be newly added commodity feature information corresponding to at least one commodity in a plurality of commodities in the model data information obtained from the blockchain. In some embodiments, a knowledge graph is constructed according to model data information and model data increment information obtained from a blockchain, the knowledge graph includes a plurality of nodes, each node can be a commodity or an associated object associated with the commodity in the model data information obtained from the blockchain, or can be a newly added commodity and an associated object associated with the newly added commodity in the model data increment information, and commodity feature information or object feature information is an attribute of a corresponding commodity node or an associated object node. In some embodiments, the knowledge graph can reflect the relationship between the nodes, in the knowledge graph, through the respective attributes of the two nodes (namely, the commodity characteristic information corresponding to the commodity node or the object characteristic information corresponding to the associated object node), the connection between the two nodes (namely, the association between the commodity and the commodity, the association between the commodity and the associated object, the association between the associated object and the associated object) can be established, the two nodes can be directly connected, or can be indirectly connected through one or more other nodes, for example, the node corresponding to the commodity A is directly connected with the node corresponding to the associated object B and is indirectly connected with the node corresponding to the commodity C through the node corresponding to the associated object B, and for example, the commodity characteristic information of the commodity A comprises "once purchased by the user U", the object characteristic information of the user B comprises "once browsed the commodity B", thereby the direct association between the commodity A and the user U can be established through the knowledge graph, the direct association between the commodity B and the user U, the indirect association between the commodity A and the commodity B (namely, the node corresponding to the commodity A is directly connected with the node corresponding to the user U, and the node corresponding to the commodity B is indirectly connected with the node corresponding to the commodity C through the node corresponding to the user U.

And the three modules 13 are used for determining second commodity calibration information according to the model data increment information and the first commodity calibration information, and updating the commodity detection model according to the model data information, the model data increment information, the connection relation characteristic information and the second commodity calibration information. In some embodiments, if the model data increment information includes at least one commodity newly added and commodity feature information corresponding to the at least one commodity, which are obtained from the blockchain, the commodity calibration information corresponding to the at least one newly added commodity is required to be obtained, and the second commodity calibration information corresponding to all the commodities in the knowledge graph is determined according to the first commodity calibration information corresponding to the model data information. In some embodiments, if the model data increment information includes newly-added commodity feature information corresponding to at least one commodity in the plurality of commodities in the model data information obtained from the blockchain, determining whether the at least one commodity has a commodity needing to be recalibrated, if so, reacquiring commodity calibration information corresponding to the part of the commodity needing to be recalibrated, and determining second commodity calibration information corresponding to all the commodities in the knowledge graph according to the first commodity calibration information corresponding to the model data information.

And the four modules 14 are used for determining model data updating information which needs to be uploaded to the blockchain according to the model data increment information, and uploading the updated commodity detection model and the model data updating information to the blockchain. In some embodiments, if the model data incremental information includes at least one commodity and commodity feature information corresponding to the at least one commodity that are newly added in addition to the model data information obtained from the blockchain, it may be determined that the model data update information that needs to be uploaded to the blockchain may include only object feature information corresponding to an associated object associated with the commodity feature information and commodity calibration information corresponding to the at least one commodity by the commodity feature information corresponding to the at least one commodity, and may also include commodity feature information corresponding to all commodities in the knowledge graph, all object feature information corresponding to all associated objects associated with all commodity feature information, and all commodity calibration information corresponding to all commodities. In some embodiments, if the model data incremental information includes newly-added commodity feature information corresponding to at least one commodity in the plurality of commodities in the model data information obtained from the blockchain, it may be determined that the model data update information to be uploaded to the blockchain may include only the newly-added commodity feature information corresponding to the at least one commodity and object feature information corresponding to an associated object associated with the newly-added commodity feature information, and commodity calibration information corresponding to the part of commodities to be recalibrated in the at least one commodity, and may also include commodity feature information corresponding to all commodities in the knowledge graph, and all object feature information corresponding to all associated objects associated with all commodity feature information, and commodity calibration information corresponding to all commodities.

In some embodiments, the apparatus is further configured to construct the knowledge-graph based on the model data information and the model data delta information. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the model data information further includes original connection relation feature information corresponding to the commodity detection model, the original connection relation feature information is obtained according to an original knowledge graph constructed by the commodity feature information and the object feature information, and the constructing the knowledge graph according to the model data information and the model data increment information includes constructing the knowledge graph according to the model data information, the model data increment information and the original connection relation feature information. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the model data increment information comprises commodity feature increment information and second object feature information corresponding to a second associated object associated with the commodity feature increment information, wherein the knowledge graph is constructed according to the model data information and the model data increment information, and the knowledge graph is constructed according to the model data information, the commodity feature increment information and the second object feature information. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the device is further configured to determine a second associated object associated with the commodity feature increment information according to the commodity feature increment information, and obtain second object feature information corresponding to the second associated object. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the commodity feature delta information comprises one or more objects, wherein the determining the second associated object with which the commodity feature delta information is associated comprises using at least one object of the one or more objects as the second associated object with which the commodity feature delta information is associated. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the associating at least one of the one or more objects as the second associated object with the merchandise feature delta information includes determining at least one object from the one or more objects and associating the at least one object as the second associated object with the merchandise feature delta information. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the determining at least one object from the one or more objects includes determining at least one object from the one or more objects based on a specific gravity of each of the one or more objects in the commodity feature delta information, wherein the specific gravity of each of the at least one object in the commodity feature delta information meets a predetermined specific gravity threshold. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the apparatus is further configured to determine a specific gravity of each object in the commodity feature delta information based on a number of occurrences of the object in the commodity feature delta information. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the apparatus further comprises determining the specific gravity of each object in the commodity feature delta information based on the semantic importance of the object in the commodity feature delta information. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the commodity feature increment information comprises first commodity feature information corresponding to at least one newly added commodity, wherein the knowledge graph is constructed according to the model data information, the commodity feature increment information and the second object feature information, and the knowledge graph is constructed according to the model data information, the first commodity feature information and the second object feature information, and the knowledge graph further comprises nodes corresponding to the at least one newly added commodity and nodes corresponding to the second associated object. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the determining the second commodity calibration information according to the model data increment information and the first commodity calibration information includes obtaining new commodity calibration information corresponding to the at least one newly added commodity, and determining the second commodity calibration information according to the new commodity calibration information and the first commodity calibration information. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the commodity feature increment information comprises newly-added commodity feature information corresponding to at least one commodity in the plurality of commodities, wherein the knowledge graph is constructed according to the model data information, the commodity feature increment information and the second object feature information, and the knowledge graph is constructed according to the model data information, the newly-added commodity feature information and the second object feature information, and the knowledge graph further comprises nodes corresponding to the second associated objects. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the determining the second commodity calibration information according to the model data increment information and the first commodity calibration information includes determining whether a commodity needing to be recalibrated exists in at least one commodity of the plurality of commodities, if yes, obtaining latest commodity calibration information corresponding to the commodity needing to be recalibrated, determining the second commodity calibration information according to the latest commodity calibration information and the first commodity calibration information, and if not, directly taking the first commodity calibration information as the second commodity calibration information. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, determining the model data update information to be uploaded to the blockchain according to the model data increment information includes determining commodity calibration increment information according to the first commodity calibration information and the second commodity calibration information, and determining the model data increment information and the commodity calibration increment information as the model data update information to be uploaded to the blockchain. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the determining the model data update information to be uploaded to the blockchain according to the model data increment information includes determining the latest model data information according to the model data information and the model data increment information, and determining the latest model data information and the second commodity calibration information as the model data update information to be uploaded to the blockchain. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the model data update information further includes the connection relationship feature information. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the device is further configured to input target commodity feature information corresponding to a target detected commodity into the updated commodity detection model, and obtain commodity detection information corresponding to the target detected commodity output by the updated commodity detection model, where the commodity detection information is used to indicate whether the target detected commodity is a failed commodity. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, the device is further configured to output at least one calibrated good in a connected relationship with the target detected good if the good detection information indicates that the target detected good is a good. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

In some embodiments, if there are a plurality of calibrated failed good having a connection relationship with the target detected good, wherein the apparatus is further configured to determine at least one calibrated failed good from the plurality of calibrated failed good, wherein a connection hop count corresponding to the connection relationship between each of the at least one calibrated failed good and the target detected good is less than or equal to a predetermined hop count threshold. The related operations are the same as or similar to those of the embodiment shown in fig. 1, and thus are not described in detail herein, and are incorporated by reference.

FIG. 5 illustrates an exemplary system that can be used to implement various embodiments described in the present application.

In some embodiments, as shown in fig. 5, the system 300 can function as any of the devices of the various described embodiments. In some embodiments, system 300 may include one or more computer-readable media (e.g., system memory or NVM/storage 320) having instructions and one or more processors (e.g., processor(s) 305) coupled with the one or more computer-readable media and configured to execute the instructions to implement the modules to perform the actions described in the present application.

For one embodiment, the system control module 310 may include any suitable interface controller to provide any suitable interface to at least one of the processor(s) 305 and/or any suitable device or component in communication with the system control module 310.

The system control module 310 may include a memory controller module 330 to provide an interface to the system memory 315. Memory controller module 330 may be a hardware module, a software module, and/or a firmware module.

The system memory 315 may be used, for example, to load and store data and/or instructions for the system 300. For one embodiment, system memory 315 may include any suitable volatile memory, such as, for example, a suitable DRAM. In some embodiments, the system memory 315 may comprise a double data rate type four synchronous dynamic random access memory (DDR 4 SDRAM).

For one embodiment, system control module 310 may include one or more input/output (I/O) controllers to provide an interface to NVM/storage 320 and communication interface(s) 325.

For example, NVM/storage 320 may be used to store data and/or instructions. NVM/storage 320 may include any suitable nonvolatile memory (e.g., flash memory) and/or may include any suitable nonvolatile storage device(s) (e.g., one or more Hard Disk Drives (HDDs), one or more Compact Disc (CD) drives, and/or one or more Digital Versatile Disc (DVD) drives).

NVM/storage 320 may include storage resources that are physically part of the device on which system 300 is installed or which may be accessed by the device without being part of the device. For example, NVM/storage 320 may be accessed over a network via communication interface(s) 325.

Communication interface(s) 325 may provide an interface for system 300 to communicate over one or more networks and/or with any other suitable device. The system 300 may wirelessly communicate with one or more components of a wireless network in accordance with any of one or more wireless network standards and/or protocols.

For one embodiment, at least one of the processor(s) 305 may be packaged together with logic of one or more controllers (e.g., memory controller module 330) of the system control module 310. For one embodiment, at least one of the processor(s) 305 may be packaged together with logic of one or more controllers of the system control module 310 to form a System In Package (SiP). For one embodiment, at least one of the processor(s) 305 may be integrated on the same die as logic of one or more controllers of the system control module 310. For one embodiment, at least one of the processor(s) 305 may be integrated on the same die with logic of one or more controllers of the system control module 310 to form a system on chip (SoC).

In various embodiments, system 300 may be, but is not limited to being, a server, workstation, desktop computing device, or mobile computing device (e.g., a laptop computing device, a handheld computing device, a tablet, a netbook, etc.). In various embodiments, system 300 may have more or fewer components and/or different architectures. For example, in some embodiments, system 300 includes one or more cameras, keyboards, liquid Crystal Display (LCD) screens (including touch screen displays), non-volatile memory ports, multiple antennas, graphics chips, application Specific Integrated Circuits (ASICs), and speakers.

The application also provides a computer readable storage medium storing computer code which, when executed, performs a method as claimed in any preceding claim.

The application also provides a computer program product which, when executed by a computer device, performs a method as claimed in any preceding claim.

The present application also provides a computer device comprising:

One or more processors;

A memory for storing one or more computer programs;

the one or more computer programs, when executed by the one or more processors, cause the one or more processors to implement the method of any preceding claim.

It should be noted that the present application may be implemented in software and/or a combination of software and hardware, e.g., using Application Specific Integrated Circuits (ASIC), a general purpose computer or any other similar hardware device. In one embodiment, the software program of the present application may be executed by a processor to perform the steps or functions described above. Likewise, the software programs of the present application (including associated data structures) may be stored on a computer readable recording medium, such as RAM memory, magnetic or optical drive or diskette and the like. In addition, some steps or functions of the present application may be implemented in hardware, for example, as circuitry that cooperates with the processor to perform various steps or functions.

Furthermore, portions of the present application may be implemented as a computer program product, such as computer program instructions, which when executed by a computer, may invoke or provide methods and/or techniques in accordance with the present application by way of operation of the computer. Those skilled in the art will appreciate that the existence of computer program instructions in a computer-readable medium includes, but is not limited to, source files, executable files, installation package files, and the like, and accordingly, the manner in which computer program instructions are executed by a computer includes, but is not limited to, the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled programs, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed programs. Herein, a computer-readable medium may be any available computer-readable storage medium or communication medium that can be accessed by a computer.

Communication media includes media whereby a communication signal containing, for example, computer readable instructions, data structures, program modules, or other data, is transferred from one system to another. Communication media may include conductive transmission media such as electrical cables and wires (e.g., optical fibers, coaxial, etc.) and wireless (non-conductive transmission) media capable of transmitting energy waves, such as acoustic, electromagnetic, RF, microwave, and infrared. Computer readable instructions, data structures, program modules, or other data may be embodied as a modulated data signal, for example, in a wireless medium, such as a carrier wave or similar mechanism, such as that embodied as part of spread spectrum technology. The term "modulated data signal" means a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. The modulation may be analog, digital or hybrid modulation techniques.

By way of example, and not limitation, computer-readable storage media may include volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data. For example, computer-readable storage media include, but are not limited to, volatile memory such as random access memory (RAM, DRAM, SRAM), and non-volatile memory such as flash memory, various read-only memory (ROM, PROM, EPROM, EEPROM), magnetic and ferromagnetic/ferroelectric memory (MRAM, feRAM), and magnetic and optical storage devices (hard disk, tape, CD, DVD), or other now known or later developed media capable of storing computer-readable information/data for use by a computer system.

An embodiment according to the application comprises an apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, trigger the apparatus to operate a method and/or a solution according to the embodiments of the application as described above.

It will be evident to those skilled in the art that the application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. A plurality of units or means recited in the apparatus claims can also be implemented by means of one unit or means in software or hardware. The terms first, second, etc. are used to denote a name, but not any particular order.

Claims

1. A method for detecting substandard goods based on blockchain, wherein the method comprises:

Send a model acquisition request to a blockchain, and receive a commodity detection model returned by the blockchain according to the model acquisition request, as well as model data information corresponding to the commodity detection model and first commodity calibration information, wherein the model data information includes commodity feature information corresponding to a plurality of commodities and object feature information corresponding to an associated object associated with the commodity feature information, the first commodity calibration information is used to calibrate whether each of the plurality of commodities is an unqualified commodity, the commodity detection model is obtained through training based on the model data information and the first commodity calibration information, and the associated object includes an object included in the commodity feature information or an object associated with the semantic content of the commodity feature information;

Obtaining connection relationship feature information corresponding to a knowledge graph, wherein the knowledge graph is constructed based on the model data information and the model data incremental information, the knowledge graph includes multiple nodes, each node in the knowledge graph corresponds to a target object in the model data information or the model data incremental information, the target object is a product or an associated object, the connection relationship feature information is used to characterize the connection relationship between the nodes in the knowledge graph, the model data incremental information includes product feature incremental information and second object feature information corresponding to a second associated object associated with the product feature incremental information, the product feature incremental information includes product feature information corresponding to at least one newly added product or newly added product feature information corresponding to at least one of the multiple products in the model data information;

Determine second commodity calibration information according to the model data increment information and the first commodity calibration information, and update the commodity detection model according to the model data information, the model data increment information, the connection relationship feature information and the second commodity calibration information, wherein the second commodity calibration information is used to calibrate all commodities in the knowledge graph;

According to the model data incremental information, the model data update information that needs to be uploaded to the blockchain is determined, and the updated commodity detection model and the model data update information are uploaded to the blockchain.

2. The method according to claim 1, wherein the step of obtaining the connection relationship feature information corresponding to the knowledge graph further comprises:

The knowledge graph is constructed according to the model data information and the model data incremental information.

3. The method according to claim 2, wherein the model data information further includes original connection relationship feature information corresponding to the commodity detection model, and the original connection relationship feature information is obtained by constructing an original knowledge graph based on the commodity feature information and the object feature information;

Wherein, constructing the knowledge graph according to the model data information and the model data increment information includes:

The knowledge graph is constructed according to the model data information, the model data incremental information and the original connection relationship feature information.

4. The method according to claim 2, wherein the model data incremental information includes commodity feature incremental information and second object feature information corresponding to a second associated object associated with the commodity feature incremental information;

Wherein, constructing a knowledge graph according to the model data information and the model data increment information includes:

The knowledge graph is constructed according to the model data information, the product feature increment information and the second object feature information.

5. The method according to claim 4, wherein the method further comprises:

According to the commodity feature increment information, a second associated object associated with the commodity feature increment information is determined, and second object feature information corresponding to the second associated object is obtained.

6. The method according to claim 5, wherein the incremental information of commodity characteristics includes one or more objects;

Wherein, the determining the second associated object associated with the incremental feature information of the commodity includes:

At least one of the one or more objects is used as a second associated object associated with the commodity feature incremental information.

7. The method according to claim 6, wherein the step of using at least one of the one or more objects as a second associated object associated with the incremental feature information of the commodity comprises:

At least one object is determined from the one or more objects, and the at least one object is used as a second associated object associated with the commodity feature incremental information.

8. The method according to claim 7, wherein the determining at least one object from the one or more objects comprises:

According to the proportion of each of the one or more objects in the product feature incremental information, at least one object is determined from the one or more objects, wherein the proportion of each of the at least one object in the product feature incremental information satisfies a predetermined proportion threshold.

9. The method according to claim 8, wherein the method further comprises:

According to the number of occurrences of each object in the incremental information of product features, the proportion of the object in the incremental information of product features is determined.

10. The method according to claim 8, wherein the method further comprises:

According to the semantic importance of each object in the incremental information of commodity features, the proportion of the object in the incremental information of commodity features is determined.

11. The method according to claim 4, wherein the incremental commodity feature information comprises first commodity feature information corresponding to at least one newly added commodity;

The step of constructing the knowledge graph according to the model data information, the incremental information of the commodity features and the second object feature information includes:

The knowledge graph is constructed according to the model data information, the first product feature information and the second object feature information, wherein the knowledge graph also includes a node corresponding to the at least one newly added product and a node corresponding to the second associated object.

12. The method according to claim 11, wherein determining the second commodity calibration information according to the model data increment information and the first commodity calibration information comprises:

The newly added commodity calibration information corresponding to the at least one newly added commodity is obtained, and the second commodity calibration information is determined according to the newly added commodity calibration information and the first commodity calibration information.

13. The method according to claim 4, wherein the incremental commodity feature information comprises newly added commodity feature information corresponding to at least one commodity among the plurality of commodities;

The knowledge graph is constructed according to the model data information, the newly added product feature information and the second object feature information, wherein the knowledge graph also includes a node corresponding to the second associated object.

14. The method according to claim 13, wherein determining the second commodity calibration information according to the model data increment information and the first commodity calibration information comprises:

Determine whether there is a commodity that needs to be recalibrated among at least one of the multiple commodities; if so, obtain the latest commodity calibration information corresponding to the commodity that needs to be recalibrated, and determine the second commodity calibration information based on the latest commodity calibration information and the first commodity calibration information; otherwise, directly use the first commodity calibration information as the second commodity calibration information.

15. The method according to claim 1, wherein determining the model data update information to be uploaded to the blockchain according to the model data incremental information comprises:

The model data incremental information and the commodity calibration incremental information are determined as the model data update information that needs to be uploaded to the blockchain.

16. The method according to claim 1, wherein determining the model data update information to be uploaded to the blockchain according to the model data incremental information comprises:

Determining the latest model data information according to the model data information and the model data increment information;

The latest model data information and the second product calibration information are determined as model data update information that needs to be uploaded to the blockchain.

The method according to claim 15 , wherein the model data update information further includes the connection relationship feature information.

The method according to claim 16 , wherein the model data update information further includes the connection relationship feature information.

19. The method according to any one of claims 1 to 18, wherein the method further comprises:

The target commodity feature information corresponding to the target detection commodity is input into the updated commodity detection model to obtain commodity detection information corresponding to the target detection commodity output by the updated commodity detection model, wherein the commodity detection information is used to indicate whether the target detection commodity is an unqualified commodity.

20. A device for detecting substandard goods based on blockchain, wherein the device comprises:

Processor; and

A memory arranged to store computer executable instructions which, when executed, cause the processor to perform a method as claimed in any one of claims 1 to 19.

21. A computer readable medium storing instructions which, when executed, cause a system to perform the operations of the method of any one of claims 1 to 19.